A Heterogeneous Ensemble Learning Framework for Spam Detection in Social Networks with Imbalanced Data

Abstract

The popularity of social networks provides people with many conveniences, but their rapid growth has also attracted many attackers. In recent years, the malicious behavior of social network spammers has seriously threatened the information security of ordinary users. To reduce this threat, many researchers have mined the behavior characteristics of spammers and have obtained good results by applying machine learning algorithms to identify spammers in social networks. However, most of these studies overlook class imbalance situations that exist in real world data. In this paper, we propose a heterogeneous stacking-based ensemble learning framework to ameliorate the impact of class imbalance on spam detection in social networks. The proposed framework consists of two main components, a base module and a combining module. In the base module, we adopt six different base classifiers and utilize this classifier diversity to construct new ensemble input members. In the combination module, we introduce cost sensitive learning into deep neural network training. By setting different costs for misclassification and dynamically adjusting the weights of the prediction results of the base classifiers, we can integrate the input members and aggregate the classification results. The experimental results show that our framework effectively improves the spam detection rate on imbalanced datasets.

Tweet-Based Bot Detection Using Big Data Analytics

“…To deal with the problem of class imbalance of spam detection in social networks, Zhao et al [68] proposed a heterogeneous stacking-based ensemble learning framework, which consists of two main modules: a base module and a combined module. In the base module, they trained six separate base classifiers to generate meta-data with new features, which are fed to the combined module.…”

Deep learning framework for handling concept drift and class imbalanced complex decision-making on streaming data

“…Table 5 and Fig. 14 computes a detailed comparative results analysis of the CIDD-ADODNN model on the test Spam dataset [23][24][25]. The resultant scores reported that HELF and KNN models have depicted inferior performance by obtaining lower accuracy values of 0.750 and 0.818, respectively.…”

Cost-Based Heterogeneous Learning Framework for Real-Time Spam Detection in Social Networks With Expert Decisions

“…To prevent misclassifications, which can be fatal to the filter, we use a cost-based machine learning technique. This method sets a different cost for errors that occur in the case of misclassification and attempts to minimize the sum of costs [43,44]. The asymmetric classification cost matrix is presented in Table II.…”

“…If is a resample that has examples, is a model generated by applying the machine learning algorithm to . Here, the risk that results when belongs to class can be defined as follows [44]:…”